By: Dan Fay (firstname.lastname@example.org), August 21, 2013 1:43 pm
Room: Moderated Discussions
> If you know the reasons which are giving rise to changes that are happening in
> CPU+OS+languages then you know that in the general picture there is no necessity
> for ARM64 because the underlying problem can be solved in a different way.
Atom, do you have any performance numbers comparing your 32-bit, multi-process model to one with a 64-bit, flat address space? If it's really that great of an idea, why not apply the framework to an application running as a whole bunch of x32 processes? If the technique is so good, you should at least see some memory savings from using the smaller, 32-bit pointers.
I also get nervous when people advocate large-scale software infrastructure changes. Telling people that they have to learn and use specific programming languages to be able to employ large data structures seems a little excessive when you can instead upgrade some relatively low-cost hardware and OS software.